Long-Term Vegetation Changes in the Amboseli Basin

Introduction

Amboseli became world renowned in the 1950s as the setting for Where No Vulture’s Fly, a film of the struggles to create Kenya’s national parks.  Famous long-horn rhinos, large-tusked elephants, teeming herds of wildlife and elegant yellow fever trees set against the background of Kilimanjaro, Amboseli drew visitors from around the world. Then, in the mid-1950s, the fever trees began dying. Conservationists blamed the Maasai for overgrazing Amboseli and pushed government to create a national park.

The Amboseli Conservation Program (ACP) began an ecological study of Amboseli in 1967, focusing on wildlife migrations and the dying woodlands. The program mapped twenty-eight distinctive vegetation zones as a baseline for monitoring future changes. The changes were mapped every five years or so in the ensuring six decades. The program set up permanent plots in the mid-1970s to monitor pasture conditions and seasonal changes in species composition of trees, shrubs, herbs and grasses. Details of vegetation mapping and monitoring methods can be found in technical reports published by the ACP team.

The study on the woodland die-off exonerated the Maasai as the cause and initially implicated a rising water table4. Later long-term exclosure experiments showed elephants alone to be the cause.  The studies showed the woodlands changes to be symptomatic of far larger ecological changes underway in Amboseli.

The studies showed plant diversity and productivity have declined, biomass turning over faster, and ecological resilience declining. Human activity has now overtaken rainfall in driving the seasonal rhythms and decadal fluctuations in plants, livestock and wildlife.

The aim of the ACP bulletins is to produce timely information on the current status and ecological changes in Amboseli for use in planning and management of the Amboseli ecosystem and national park. This bulletin updates earlier publications on the long-term changes in vegetation and the underlying causes. We give vegetation trends and present results in graphic form for ease of viewing. We conclude with comment on the causes of change and the implications for the management of Amboseli.

Amboseli Basin Vegetation Changes 1950-2023. Noticeable changes include the reduction of dense bushlands and woodlands, an expansion of grasslands and open bushlands, and fluctuations in the size of open water, permanent swamps and swamp edges.

The once extensive fever tree woodlands which extended the length of the Amboseli Basin in the 1950 have been replaced by Suaeda shrublands due to the heavy browsing.

Our Team

Ms. Immaculate
Ombongi

Data Analyst

Immaculate is a data analyst at ACP. She has a Bachelors’ degree in Financial Economics from Mount Kenya University.

She is experienced in spatial data analysis and modeling of livestock markets in Kenya. Her interests include GIS, remote sensing, satellite imagery processing and analysis.

Immaculate as well, supports the analysis team that is working on the Rangeland restoration, a program of the African Conservation Centre, also known as the JUSTDIGGIT project.

Our Team

Ms. Immaculate
Ombongi

Data Analyst

Immaculate is a data analyst at ACP. She has a Bachelors’ degree in Financial Economics from Mount Kenya University.

She is experienced in  spatial  data analysis and modeling of  livestock markets in Kenya.  Her interests include GIS, remote sensing, satellite imagery processing and analysis.

Immaculate as well, supports the analysis  team  that is working on the Rangeland restoration, a program of the African Conservation Centre, also known as the JUSTDIGGIT project.

Our Team

Mr. Sakimba
Kimiti

Assistant Researcher

Sakimba is currently pursuing a PhD at the University of Lyon 2 in France. He previously worked as an Assistant Researcher for the Amboseli Conservation Program.

He holds a Bachelor of Science (Wildlife Management and Conservation) degree from the University of Nairobi and  a Master of Science degree in Range Management from the same University.

Prior to joining the ACP, he worked as an Ecological Assistant at South Rift Land Owners Association. At ACP, he is involved in projects dealing with the Dynamics of Predation on Spatial -temporal Basis and in Human Ecology.

His other interests include: GIS, remote sensing, satellite imagery, ecological monitoring, land use change and ecosystem vulnerability.

Our Team

Mr. Sakimba
Kimiti

Assistant Researcher

Sakimba is currently pursuing a PhD at the University of Lyon 2 in France. He previously worked as an Assistant Researcher for the Amboseli Conservation Program.

He holds a Bachelor of Science (Wildlife Management and Conservation) degree from the University of Nairobi and  a Master of Science degree in Range Management from the same University.

Prior to joining the ACP, he worked as an Ecological Assistant at South Rift Land Owners Association. At ACP, he is involved in projects dealing with the Dynamics of Predation on Spatial -temporal Basis and in Human Ecology.

His other interests include: GIS, remote sensing, satellite imagery, ecological monitoring, land use change and ecosystem vulnerability.

Our Team

Ms. Winfridah
Kemunto

Database Administrator

Winfridah  is the  Amboseli Conservation Program’s database Administrator. She has a certificate from Pitman Training Institute and vast experience in working with big data that involve database management,  basic analysis, digital library, data mining and  data visualization.

Her interests include spatial data mining and presentation.  Before Joining ACP, she worked  as a data clerk at South Rift Land Owners Association (SORALO).

Our Team

Ms. Winfridah
Kemunto

Database Administrator

Winfridah  is the  Amboseli Conservation Program’s database Administrator. She has a certificate from Pitman Training Institute and vast experience in working with big data that involve database management,  basic analysis, digital library, data mining and  data visualization.

Her interests include spatial data mining and presentation.  Before Joining ACP, she worked  as a data clerk at South Rift Land Owners Association (SORALO).

Our Team

Mr. David
Maitumo

Field Officer/ Data Collector

David has been working in Amboseli as the ACP field officer since 1977. As a member of the local Maasai community in the Amboseli area, David brings a unique perspective to the program.

His rich understanding of the interaction of people, livestock, and wildlife, and the challenges facing conservation in human landscapes, enriches his key roles in the design of field experiments and long term data collection and monitoring.

Our Team

Mr. David
Maitumo

Field Officer/ Data Collector

David has been working in Amboseli as the ACP field officer since 1977. As a member of the local Maasai community in the Amboseli area, David brings a unique perspective to the program.

His rich understanding of the interaction of people, livestock, and wildlife, and the challenges facing conservation in human landscapes, enriches his key roles in the design of field experiments and long term data collection and monitoring.

Our Team

Dr. David
Western

Founder & Chairman

Dr. David Western, known as Jonah, began research into savannas ecosystems at Amboseli in 1967, looking at the interactions of humans and wildlife.

His work, unbroken since then, has served as a barometer of changes in the savannas and test of conservation solutions based on the continued coexistence of people and wildlife.

Jonah is currently chairman of the African Conservation Centre, Nairobi. He directed Wildlife Conservation Society programs internationally, established Kenya’s Wildlife Planning Unit, chaired the World Conservation Union’s African Elephant and Rhino Specialist Group, and was founding president of The International Ecotourism Society, chairman of the Wildlife Clubs of Kenya, director of Kenya Wildlife Service, and founder of the African Conservation Centre in Nairobi.

He is an adjunct professor in Biology at the University of California, San Diego.

Western’s publications include;

Conservation for the Twenty-first Century (OUP, 1989), Natural Connections: Perspectives in Community-based Conservation (Island Press, 1994) and In the Dust of Kilimanjaro (Shearwater, 2001).

He is presently conducting a study on climate change in the Kenya-Tanzania borderlands in collaboration with University of California San Diego, University of York, Missouri Botanical Gardens, and African Conservation Centre.

Our Team

Dr. David
Western

Founder & Chairman

Dr. David Western, known as Jonah, began research into savannas ecosystems at Amboseli in 1967, looking at the interactions of humans and wildlife.

His work, unbroken since then, has served as a barometer of changes in the savannas and test of conservation solutions based on the continued coexistence of people and wildlife.

Jonah is currently chairman of the African Conservation Centre, Nairobi. He directed Wildlife Conservation Society programs internationally, established Kenya’s Wildlife Planning Unit, chaired the World Conservation Union’s African Elephant and Rhino Specialist Group, and was founding president of The International Ecotourism Society, chairman of the Wildlife Clubs of Kenya, director of Kenya Wildlife Service, and founder of the African Conservation Centre in Nairobi.

He is an adjunct professor in Biology at the University of California, San Diego.

Western’s publications include;

Conservation for the Twenty-first Century (OUP, 1989), Natural Connections: Perspectives in Community-based Conservation (Island Press, 1994) and In the Dust of Kilimanjaro (Shearwater, 2001).

He is presently conducting a study on climate change in the Kenya-Tanzania borderlands in collaboration with University of California San Diego, University of York, Missouri Botanical Gardens, and African Conservation Centre.

Our Team

Dr. Victor N.
Mose

Deputy Director & Head of Bio-statistical Services

Dr. Victor N. Mose is the Deputy Director and Head of Biostatistical Services. He was awarded a PhD in Biomathematics by the University of Pierre and Marie Curie (UPMC), Paris VI, France in 2013.

He has a Masters in bio-statistics from the University of Nairobi, Kenya and a Bachelors degree in Mathematics from the same University.

He also holds a financial mathematics qualification from the Institute of Actuaries, London, UK.

Victor is experienced in ecological modeling, bio-informatics, and geographical information systems (GIS).

His research interests include Population dynamics, migration modelling, Bayesian spatial analysis, ecosystem services and economics modelling, together with biodiversity mapping.

Victor’s publications include;

Mose, V.N., Nguyen-Huu, T., Auger, P., Western, D. 2012. Modelling herbivore population dynamics in the Amboseli National Park, Kenya: Application of spatial aggregation of variables to derive a master model. Ecological Complexity, 10, 42-51.

Our Team

Dr. Victor N.
Mose

Deputy Director & Head of Bio-statistical Services

Dr. Victor N. Mose is the Deputy Director and Head of Biostatistical Services. He was awarded a PhD in Biomathematics by the University of Pierre and Marie Curie (UPMC), Paris VI, France in 2013.

He has a Masters in bio-statistics from the University of Nairobi, Kenya and a Bachelors degree in Mathematics from the same University.

He also holds a financial mathematics qualification from the Institute of Actuaries, London, UK.

Victor is experienced in ecological modeling, bio-informatics, and geographical information systems (GIS).

His research interests include Population dynamics, migration modelling, Bayesian spatial analysis, ecosystem services and economics modelling, together with biodiversity mapping.

Victor’s publications include;

Mose, V.N., Nguyen-Huu, T., Auger, P., Western, D. 2012. Modelling herbivore population dynamics in the Amboseli National Park, Kenya: Application of spatial aggregation of variables to derive a master model. Ecological Complexity, 10, 42-51.

Figure 12:Illustration of how the monthly total count of keystone species is conducted within the Amboseli basin area that includes that protected Amboseli National Park. The counts are done on a one-by-one km grid system.

Figure 1: Amboseli National Park is surrounded by Maasai group ranches. ACP has conducted aerial surveys of the 8,500 square kilometers eastern Kajiado region since 1973 using a 5 x 5 kilometer-square grids to count and map wildlife and livestock. The brown box (migration area) defines the Amboseli ecosystem—the seasonal range of the migratory wildlife populations using Amboseli National Park and permanent swamps in the dry season. Ground vegetation monitoring is done at basin area surrounding the Park and in selected group ranches (Olgulului, Kimana, Eselenkei and Mbirikani).

Figure 9 : Livestock body condition scores and milk yield in the Amboseli ecosystem

Figure 8: Counting of all herbivore species (including livestock) seen over the 500m radius from the centre of the plots scattered across the Amboseli ecosystem .

Table 1: Calculations and estimates of vegetation variables collected in the permanent plots in the basin and the surrounding group ranches.

Figure 5: Percentage cover showing number of hits, misses and total number of pins captured. For instance, percentage cover here is calculated as: (Number of Hits)/(Total number of pins) × 100, which gives, 5/9 × 100=55.6. The grass height is recorded in cm.

Figure 3: Ground vegetation monitoring in selected group ranches. The plots where monitoring is done within the ranches are also shown.

Figure 2: Major habitats amalgamated into eight habitats from 29 vegetation zones defined in the 1967 baseline and subsequent surveys. The 10m radius permanent plots monitored every 4 to 6 weeks are shown. The 20 vegetation plots were selected from a randomized set of 101 original plots. (4)

Table 4: Some satellites and their Spatio-temporal resolution

Figure 15: A screenshot of Google Earth showing human settlement locations in the Amboseli basin area. The red placemarks show occupied settlements while the yellow ones represent the unoccupied

Figure 16: Detected bomas identified by the use of Machine Learning in the Amboseli basin area.

Figure 17: A recent sample of the Sentinel 2 image obtained for processing of the vegetation zones of the Amboseli basin area. The swamp habitat is reasonably distinguished by color red.

Figure 18: Historical changes in the Amboseli basin vegetation from aerial photography mapping that are currently being updated using satellite imagery.